English

Combinatorics of the symmetries of ascents in restricted inversion sequences

Combinatorics 2021-12-09 v1

Abstract

The systematic study of inversion sequences avoiding triples of relations was initiated by Martinez and Savage. For a triple (ρ1,ρ2,ρ3){<,>,,,=,,}3(\rho_1,\rho_2,\rho_3)\in\{<,>,\leq,\geq,=,\neq,-\}^3, they introduced \In(ρ1,ρ2,ρ3)\I_n(\rho_1,\rho_2,\rho_3) as the set of inversion sequences e=e1e2ene=e_1e_2\cdots e_n of length nn such that there are no indices 1i<j<kn1\leq i<j<k\leq n with eiρ1eje_i \rho_1 e_j, ejρ2eke_j \rho_2 e_k and eiρ3eke_i \rho_3 e_k. To solve a conjecture of Martinez and Savage, Lin constructed a bijection between \In(,,>)\I_n(\geq,\neq,>) and \In(>,,)\I_n(>,\neq,\geq) that preserves the distinct entries and further posed a symmetry conjecture of ascents on these two classes of restricted inversion sequences. Concerning Lin's symmetry conjecture, an algebraic proof using the kernel method was recently provided by Andrews and Chern, but a bijective proof still remains mysterious. The goal of this article is to establish bijectively both Lin's symmetry conjecture and the γ\gamma-positivity of the ascent polynomial on \In(>,,>)\I_n(>,\neq,>). The latter result implies that the distribution of ascents on \In(>,,>)\I_n(>,\neq,>) is symmetric and unimodal.

Keywords

Cite

@article{arxiv.2112.04115,
  title  = {Combinatorics of the symmetries of ascents in restricted inversion sequences},
  author = {Joanna N. Chen and Zhicong Lin},
  journal= {arXiv preprint arXiv:2112.04115},
  year   = {2021}
}

Comments

23 pages, 2 figures